Lipid phosphate phosphatase inhibitors locally amplify lysophosphatidic acid LPA1 receptor signalling in rat brain cryosections without affecting global LPA degradation
- 6.4k Downloads
Lysophosphatidic acid (LPA) is a signalling phospholipid with multiple biological functions, mainly mediated through specific G protein-coupled receptors. Aberrant LPA signalling is being increasingly implicated in the pathology of common human diseases, such as arteriosclerosis and cancer. The lifetime of the signalling pool of LPA is controlled by the equilibrium between synthesizing and degradative enzymatic activity. In the current study, we have characterized these enzymatic pathways in rat brain by pharmacologically manipulating the enzymatic machinery required for LPA degradation.
In rat brain cryosections, the lifetime of bioactive LPA was found to be controlled by Mg2+-independent, N-ethylmaleimide-insensitive phosphatase activity, attributed to lipid phosphate phosphatases (LPPs). Pharmacological inhibition of this LPP activity amplified LPA1 receptor signalling, as revealed using functional autoradiography. Although two LPP inhibitors, sodium orthovanadate and propranolol, locally amplified receptor responses, they did not affect global brain LPA phosphatase activity (also attributed to Mg2+-independent, N-ethylmaleimide-insensitive phosphatases), as confirmed by Pi determination and by LC/MS/MS. Interestingly, the phosphate analog, aluminium fluoride (AlFx-) not only irreversibly inhibited LPP activity thereby potentiating LPA1 receptor responses, but also totally prevented LPA degradation, however this latter effect was not essential in order to observe AlFx--dependent potentiation of receptor signalling.
We conclude that vanadate- and propranolol-sensitive LPP activity locally guards the signalling pool of LPA whereas the majority of brain LPA phosphatase activity is attributed to LPP-like enzymatic activity which, like LPP activity, is sensitive to AlFx- but resistant to the LPP inhibitors, vanadate and propranolol.
KeywordsPropranolol Brain Section Phosphatidic Acid Nadolol LPA1 Receptor
Bovine serum albumin
G protein-coupled receptor
Liquid chromatography/tandem mass spectrometry
Lipid phosphate phosphatase
Lysophosphatidic acid (LPA, 1- or 2-acyl-sn-glycero-3-phosphate) is a signalling phospholipid mediating multiple biological responses, such as cellular proliferation, prevention of apoptosis, and platelet aggregation, and is involved in the development and function of the nervous, cardiovascular, immune, and reproductive systems [1, 2]. Aberrant LPA signalling has been claimed to be associated with the pathology of common human diseases, such as arteriosclerosis  and cancer . Signalling by LPA is mainly mediated through specific G protein-coupled receptors (GPCRs) [5, 6, 7]. The receptors for LPA are widely expressed, being found in the brain, circulation and digestive tract. Currently, there are five GPCRs that have been identified as bona fide receptors for LPA (LPA1-5) along with a putative sixth receptor (LPA6) .
Physiologically relevant levels of LPA can be found in serum and other body fluids. In addition, several cell types, including platelets, adipocytes, and ovarian cancer cells, can produce and release LPA. Significant amounts of LPA have been detected from brain tissue [9, 10, 11]. It is postulated that the majority of bioactive LPA is generated extracellularly from lysophospholipids, such as lysophosphatidylcholine (LPC), by the plasma ecto-enzyme, lysophospholipase D, identical to autotaxin, an autocrine motility factor originally isolated from the conditioned medium of cancer cells . Intracellularly, LPA can be generated by phospholipase A1/A2 (PLA1/PLA2, respectively) -catalyzed deacylation of phosphatidic acid (PA) [13, 14]. Other proposed pathways for LPA generation include de novo biosynthesis either from glycerol-3-phosphate (GP) by glycerol-3-phosphate acyltransferase or from monoacylglycerol (MAG) by monoacylglycerol kinase .
35 S]GTPγS autoradiography represents a powerful functional approach to anatomically localize receptor-dependent Gi/o protein activity directly in brain cryosections (for reviews, see [24, 25]). In addition to a receptor’s anatomical distribution, 35 S]GTPγS autoradiography can monitor the receptor-G protein axis in its native cellular microenvironment and is therefore often referred to as functional autoradiography. Our previous studies have demonstrated that brain sections retain the capacity to generate endogenous GPCR agonists, such as adenosine and LPA, during incubation. This results in tonic adenosine A1 and LPA receptor activity in anatomically defined brain regions and therefore serves as a convenient functional readout to monitor agonist activity at the two receptors [26, 27, 28]. The LPA-evoked 35 S]GTPγS binding response in rat brain sections reflects LPA1 receptor activity, as it is sensitive to the LPA1/3-selective antagonist Ki16425 and is restricted to the developing white matter tracts [24, 26, 29]. This labelling pattern faithfully mirrors the known expression pattern of LPA1 receptors in the developing rat brain [30, 31, 32, 33].
Our previous studies indicated that the enzymatic machinery generating and metabolizing membrane-derived lipid mediators was well preserved in brain cryosections. We recently demonstrated that a comprehensive elimination of the enzymatic hydrolysis of the endocannabinoid 2-arachidonoylglycerol (2-AG) in brain sections leads to 2-AG accumulation and subsequent cannabinoid CB1 receptor activation, as successfully revealed using functional autoradiography . Using this approach, we show here that pharmacological inhibition of LPP activity in brain sections with vanadate or propranolol results in amplification of LPA1 receptor signalling with no net effect on global LPA phosphatase activity at the bulk brain level, also attributable to Mg2+ -independent, NEM-resistant LPP-like phosphatases. We show further that the phosphate analog AlFx- not only potentiates LPA1 receptor signalling, but also totally prevents LPA degradation, resulting in the accumulation of several LPA species in brain sections, as demonstrated by LC/MS/MS measurements. The presently described approach offers a versatile tool to monitor the strength of lipid-GPCR signalling axis in anatomically defined brain structures and may prove useful also for further studies exploring enzymatic pathways estimating the lifetime of still uncharacterized endogenous signalling lipids.
The LPP inhibitors Na3VO4 and propranolol locally amplify LPA1 receptor signalling without affecting global LPA degradation
The phosphate analog aluminium fluoride amplifies LPA1 receptor signalling and totally prevents LPA degradation resulting in bulk accumulation of endogenous LPA species
We were curious to examine whether AlFx- could also inhibit LPA degradation at the bulk brain level. When sections were pretreated with AlFx- but then omitted from all subsequent steps, AlFx- readily facilitated LPA1 receptor signalling (Figure 4, Figure 5a), but such a pretreatment did not inhibit LPA degradation in a statistically significant manner. Degradation of exogenous LPA (50 μM) alone yielded 22.6 ± 1.2 nmol Pi per slide whereas pretreatment with AlFx- followed by incubation with exogenous LPA yielded 21.4 ± 0.4 nmol Pi per slide (mean ± SEM, n = 3). However, when added together with LPA, AlFx- totally (and NaF partially) blocked the formation of LPA-derived Pi, thus providing evidence of the ability of these compounds to inhibit the vanadate- and propranolol-insensitive pool of LPA phosphatases in a reversible manner (Figure 5b). Treatment with DFOM (50 μM) totally prevented the ability of AlFx- and NaF to inhibit the degradation of LPA (Figure 5b), indicating that AlFx-, rather than NaF, was the active compound.
The LPA → MAG → G pathway efficiently degrades exogenous LPA whereas the LPA → GP → G pathway is inactive
Since both of the two Pi releasing pathways for LPA degradation finally produce glycerol, we assessed cerebellar membrane-dependent glycerol generation from exogenous LPA in brain tissue. Monoglyceride lipase (MGL) is believed to be mainly responsible for the MAG → G conversion. In addition, two novel α/β-hydrolase domain containing proteins, ABHD6 and ABHD12, have been identified to hydrolyze brain endocannabinoid 2-arachidonoylglycerol (2-AG)  and together the three serine hydrolases account for ~99% of brain 2-AG hydrolase activity . It is therefore likely that in addition to MGL, ABHD6 and ABHD12 are involved in the degradation of both 1- and 2-monoacylglycerols. To delineate the relative contributions of the three hydrolases, we pretreated cerebellar membranes with two serine hydrolase inhibitors, methylarachidonoylfluorophosphonate (MAFP) and compound JZL184. The former is a potent, non-selective inhibitor of MGL [39, 42] that also inhibits ABHD6/ABHD12, whereas the latter is a MGL-selective inhibitor . As expected, in rat cerebellar membranes incubated together with LPA (10 μM), glycerol production closely matches with Pi generation (Figure 7a and b), indicating that MAG → G conversion takes place under the assay conditions employed. With MAFP pretreatment (1 μM), LPA-derived glycerol production was decreased by 91% (Figure 7b). With JZL184 pretreatment (100 μM), the corresponding reduction was 71% (Figure 7b). The selectivity of the inhibitors towards MGL likely explains the difference in the inhibition of glycerol production from LPA between the two inhibitors. The time-dependent generation of Pi and glycerol from exogenous LPA is presented in Figure 7c. It appears that Pi generation precedes that of G, a finding supporting sequential actions of phosphatases and lipases on the LPA → MAG → G pathway.
Functional autoradiography provides a straightforward approach to study the proximal step of signalling of various Gi/o-coupled receptors in brain cryosections. We recently demonstrated that brain sections retain the capacity to generate endocannabinoids during incubations i.e. evidence of the ability of the brain sections to preserve sufficient functional enzymatic machinery to generate endogenous GPCR activating ligands . The lifetime of the signalling pool of LPA is thought to be controlled by the equilibrium between synthesizing and degradative enzymatic activity. In the current study, we have characterized these enzymatic pathways and their role in tonic LPA1 receptor activity by pharmacologically manipulating the enzymatic machinery required for LPA degradation. We observed that in brain sections, the lifetime of bioactive LPA is controlled by Mg2+-independent, NEM-insensitive phosphatase activity attributable to LPPs. Pharmacological inhibition of this LPP activity by AlFx-, propranolol or sodium orthovanadate amplified LPA1 receptor signalling, as revealed using functional autoradiography. We provided further evidence to show that the majority of brain LPA phosphatase activity seems to be carried out by LPP-like enzymatic activity which like LPP activity, is sensitive to AlFx- but appears to be resistant to the two other LPP inhibitors, vanadate and propranolol. Finally, we demonstrated that degradation of exogenous LPA is almost entirely channelled via the LPA → MAG → G pathway and that MGL accounts for the majority of oleylglycerol-hydrolyzing activity in brain tissue.
All the three subtypes of Mg2+-independent/NEM resistant LPPs are expressed in the brain, yet very little is known about the role of the LPPs as regulators of LPA receptor signalling in the nervous system. Knockout studies of all the LPP subtypes have been reported [44, 45, 46]. Study with LPP1 knockout mice indicated that LPP1 plays a role in regulating the degradation of circulating LPA in vivo but that study failed to disrupt the LPP1 encoding gene in the brain, obscuring the function of LPP1 in the nervous system . Knockout of LPP3 turned out to be embryonically lethal  whereas in vitro studies using cell lines lacking LPP3 address involvement of LPP3 in early neural development . The LPPs are likely to be involved in LPA dephosphorylation in brain cryosections, as brain sections efficiently generate Pi from exogenous LPA largely in a NEM resistant and Mg2+-independent way. Propranolol and vanadate have been demonstrated to inhibit LPPs in various cell types [20, 35, 36, 48], vanadate also in the rat brain . Propranolol has been shown to act as a moderately effective inhibitor of LPPs  supporting our finding where the vanadate-induced response is relatively stronger when compared to the response observed with propranolol. Since propranolol and vanadate amplified LPA1 receptor signalling only when present in the 35 S]GTPγS labelling step, these drugs presumably inhibit LPPs in a reversible manner. In brain sections, LPP activity appears to locally control the lifetime of the signalling pool of LPA and LPPs must therefore reside in close proximity to the LPA1 receptors, as propranolol and vanadate had no effect on LPA degradation when assessed at the bulk brain level.
In functional autoradiography, AlFx- more efficiently induced the LPA1 receptor-mediated signal as compared to the signals observed with vanadate or propranolol. Since AlFx- is able to induce the LPA1 receptor-mediated signal when present only in the pre-incubation step, it appears to inhibit LPPs in an irreversible manner. This proposal is supported by the finding that the Al3+ chelator DFOM failed to reverse AlFx- -evoked response, if added only after pretreatment of brain sections with AlFx- (and NaF). AlFx- is known to mimic the chemical structure of phosphate and therefore affects the activity of several phosphoryl transfer enzymes . As a phosphate analog, AlFx- might bind to the Pi recognizing binding pocket of the LPPs and by this mechanism lead to irreversible inhibition. All the studied inhibitors evoked 35 S]GTPγS binding responses that were largely restricted to the white matter areas of the brain when compared to grey matter (See Additional file 7: Inhibitor-evoked 35 S]GTPγS binding responses are restricted to the white matter areas of the brain) reflecting to selectivity towards the myelin-enriched LPA1 receptors. This also provides evidence to show, that though AlFx- is known to act as a general activator of heterotrimeric G proteins, it seems not to induce global binding response in the grey matter areas and therefore seems not to act as a general G protein activator in functional autoradiography. It is notable that in contrast to propranolol and vanadate, when present in the latter step together with exogenous LPA, AlFx- totally prevented the degradation of LPA at the bulk brain level, suggesting that in addition to irreversibly inhibiting LPPs, AlFx- can inhibit other LPP-like phosphatases in a reversible manner. Based on the present findings, the major portion of brain LPA phosphatase activity appears to be attributable to the LPP-like phosphatases which in a similar manner as LPPs, are sensitive to AlFx- but resistant to the LPP inhibitors, vanadate and propranolol.
Since there was no Pi generation from exogenous glycerol 3-phosphate, it seems that LPA is predominantly degraded by the LPA → MAG → G pathway in our experimental setting whereas the LPA → GP → G pathway plays a minor role. According to our findings, both phosphohydrolases (LPPs/LPP-like) and MGL and related hydrolases (ABHD6/ABHD12) seem to be active. The Pi- and glycerol -generating enzymatic routes involved in LPA degradation are summarized in Additional file 8: Summary of enzymatic routes generating Pi and glycerol. Previously, NEM-insensitive LPA phosphohydrolase activity was studied in the nuclear fraction isolated from rabbit cerebral cortex . This activity was found to be present also in the microsomal fraction. In the nuclear fraction, phosphohydrolase activity was found to be sensitive to NaF (50 mM) but virtually insensitive to propranolol (0.5 mM). Dephosphorylation by phosphohydrolases was found to be more active route for LPA degradation when compared to deacylation by lysophospholipases. It was also indicated that followed by dephosphorylation of LPA, monoacyl product is rapidly converted to glycerol by monoglyceride lipase. These findings support our present findings concerning active pathways involved in LPA degradation in brain as well as about the existence of LPP-like, propranolol and vanadate -insensitive, phosphohydrolase activity.
The LPP-like phosphatases remain to be characterized in future experiments. One interesting group of brain-specific membrane proteins are plasticity related genes (PRGs) that have recently been identified and were originally proposed to act as LPA phosphatases . Among the family of PRGs, PRG-1 shares close homology to the LPPs, having three conserved integral domains facing the extracellular side of the plasma membrane, the feature that enables LPPs to dephosphorylate their lipid substrates. However, the catalytic residues responsible for LPP activity are not fully conserved in PRGs  suggesting that PRGs might not act as LPA phosphatases. Instead, PGR-1 was recently demonstrated to act at the postsynaptic side of the excitatory glutamatergic synapse where it could mediate the uptake of bioactive lipids . PRG-1 was found to effectively control the levels of LPA in the synapse though its mechanism of action seems to be more transporter-like than dephosphorylating. The transporter mechanism is not expected to be active in our experimental setting and therefore we hypothesize that PRGs are not controlling LPA levels in our model.
We demonstrate that the lifetime of bioactive LPA is controlled by LPPs in rat brain cryosections and that pharmacological inhibition of this LPP activity results in amplification of basal and LPA-stimulated LPA1 receptor activity. We conclude that LPP acts locally to control the lifetime of the signalling pool of LPA in the vicinity of LPA1 receptors whereas the majority of brain LPA phosphatase activity is attributable to additional LPP-like enzymatic activity. Functional autoradiography represents a valuable tool for studies into LPA degradation by LPPs and LPP-like enzymatic activity. Compounds affecting LPA degradation could prove to be attractive targets for drug development, since altered LPA levels are associated with common human diseases, e.g. several forms of cancer. The approach described in this paper may also prove useful for further studies elucidating enzymatic pathways regulating the lifetime of still uncharacterized endogenous signalling lipids.
Propranolol was purchased from Biomol (Plymouth Meeting, PA, USA). AlCl3 and NaF were from Merck (Darmstat, Germany). 1-oleoyl-2-methyl-sn-glycero-3-phosphothionate ((2S)-OMPT), 1-palmitoyl-2-hydroxy-sn-glycero-3-phosphate (16:0 LPA), 1-heptadecanoyl-2-hydroxy-sn-glycero-3-phosphate (17:0 LPA), 1-stearoyl-2-hydroxy-sn-glycero-3-phosphate (18:0 LPA), 1-arachidonoyl-2-hydroxy-sn-glycero-3-phosphate (20:4 LPA) were from Avanti Polar Lipids (Alabaster, AL, USA). 1-oleyl-2-hydroxy-sn-glycero-3-phosphate (18:1 LPA), sodium orthovanadate (Na3VO4), nadolol, bovine serum albumin (BSA, fatty acid free), dithiothreitol (DTT), 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), guanosine diphosphate (GDP), 3-(4-[4-([1-(2 chlorophenyl)ethoxy] carbonyl amino)-3-methyl-5-isoxazolyl]benzylsulfanyl) propanoic acid (Ki16425), GTPγS, N-ethylmaleimide (NEM), glycerol 3-phosphate, and deferoxamine mesylate (DFOM) were from Sigma (St Louis, MO, USA). Methylarachidonoylfluorophosphonate (MAFP) and compound JZL184 were from Cayman Chemical (Ann Arbor, MI, USA). [35 S]GTPγS (initial specific activity 1250 Ci/mmol) was purchased from NEN Life Science Products Inc. (Boston, MA, USA). All other chemicals were of the finest purity available.
Experiments were performed using 4-week old male Wistar rats obtained from the National Laboratory Animal Centre, University of Eastern Finland, Kuopio, Finland. Approval for the harvesting of animal tissue was applied, registered and obtained from the local welfare officer of the University of Eastern Finland. The experiments did not involve any in vivo treatment. The animals were housed in groups of five to ten individuals per cage under standard laboratory conditions (12–12 h light–dark cycle, food and water ad libitum, 60% relative humidity). The rats were decapitated 8–9 h after lights on, and within the next 5 min, the whole brain was dissected out, dipped briefly in isopentane (chilled on dry ice) and stored at −80 °C. Horizontal, coronal or sagittal brain sections (20 μm thick) were cut at −19 °C to −21 °C using a Leica cryostat, thaw-mounted onto Superfrost®Plus slides (Menzel-Gläser, Germany), dried for 1–4 h at room temperature under a constant stream of air and stored thereafter at −80 °C.
[35 S]GTPγS autoradiography
35 S]GTPγS autoradiography was performed as previously described [26, 28, 34]. Briefly, the brain sections were processed in three sequential steps consisting of pre-incubation for 20/40 min (step 1), GDP-loading for 50/60 min (step 2), and 35 S]GTPγS labelling for 90 min (step 3). For some treatments, a brief additional incubation (10 min) was performed prior to step 1, as detailed in Results. All the steps were performed at 20 °C using Tris-based buffer (50 mM Tris–HCl, pH 7.4, 1 mM EDTA, 100 mM NaCl, 5 mM MgCl2). Steps 2 and 3 included 1 μM DPCPX (to block the tonic adenosine A1 receptor signal) and 2 mM GDP and step 3 additionally included 1 mM DTT and 40–250 pM 35 S]GTPγS. Chemicals being investigated (LPA species used was 18:1 LPA) and 0.1% BSA (fatty acid free) were included in the assay during steps 1,2 or 3 as described in Results. In addition, some slides in each experiment were incubated in the presence of 10 μM GTPγS to determine non-specific binding. After the 90 min autoradiography step, the slides were rinsed twice (5 min each time) in ice-cold washing buffer (50 mM Tris–HCl, pH 7.4 and 5 mM MgCl2), dipped for 30 s in ice-cold deionized water, and air-dried. The slides were arranged into a cassette together with 14 C] standard (Amersham, Little Chalfont, Bucks, UK) and exposed against a radiosensitive film (BioMax MR™, Kodak Scientific Imaging Film) for 2–5 days. After exposure, the films were developed for 3–4 min at 4 °C with Kodak D-19 developer.
[35 S]GTPγS membrane binding assay
A previously described method was utilized for preparation of membranes . Briefly, eight cerebella were homogenized in nine volumes of ice-cold 0.32 M sucrose. The homogenate was centrifuged at 1000 × g for 10 min at 4 °C and the resulting supernatant was centrifuged at 100 000 × g for 30 min at 4 °C. The high speed centrifugation was repeated twice, resuspending the pellet in ice-cold deionised water. After the final centrifugation, the membranes were suspended in Tris–HCl (50 mM, pH 7.4) supplemented with EDTA (1 mM) and stored thereafter at −80 °C.
The 35 S]GTPγS membrane binding assay was performed as previously described [28, 34], with minor modifications. Briefly, the final incubation volume (400 μl) contained 5 μg membrane protein in the incubation buffer (50 mM Tris–HCl, pH 7.4, 1 mM EDTA, 100 mM NaCl, 5 mM MgCl2, 10 μM GDP, 1 mM DTT, 0.5% BSA, and ~ 150 pM 35 S]GTPγS) plus the chemicals of interest. The incubation buffer was supplemented with 1 μM DPCPX to suppress basal 35 S]GTPγS binding due to endogenous adenosine. The membranes were pre-incubated for 30 min prior to conducting the 35 S]GTPγS binding step. The incubations were stopped after 90 min by the addition of 4 ml ice-cold washing buffer (50 mM Tris–HCl, pH 7.4 and 5 mM MgCl2), followed by filtration through glass fibre filters (Whatman GF/B) and by two additional washes with the buffer. The filters were transferred into scintillation vials along with HiSafe3 scintillation liquid (Wallac, Turku, Finland). After vertical shaking for 15 min to extract radioactivity trapped in filters, the tubes were counted the next day with Wallac LKB 1213 Rackbeta, Wallac, Turku, Finland.
Determination of Pi and glycerol
To estimate LPA-degrading phosphatase activity under the assay conditions mimicking functional autoradiography, triplicate slides with two or three horizontal brain sections underwent the autoradiography mimicking protocol (except that DPCPX, GDP and radioligand were omitted). The experiments with Na3VO4, propranolol, and nadolol (as a negative control for propranolol) were performed with and without NEM pretreatment. The protocol consisted of 10 min pre-incubation in the assay buffer, 30 min incubation in the assay buffer with or without 5 mM NEM, then a washing step with assay buffer, and finally 90 min incubation in the presence of chemicals in interest as well as 0.1% BSA and 1 mM DTT. In NEM-treated sections, Mg2+ was omitted from the assay buffer. In experiments with NaF and AlFx-, the protocol consisted of two sequential 40 min incubations (steps 1 and 2) and finally 90 min incubation (step 3) in the presence of the chemicals of interest as well as 0.1% BSA and 1 mM DTT. LPA species used was 18:1 LPA. In all experiments, after the final 90 min incubation step, the postincubation buffer was collected quantitatively and the Pi content was determined in duplicate using the Fiske-Subbarow method, as described in Esmann (1988)  after TCA-precipitation of BSA which interfered with the Pi determinations. Absorbances (λ 700 nm) were read with a Tecan Spectrafluor microplate reader.
To clarify the enzymatic routes responsible for LPA degradation in our experimental setting, Pi formation was determined using enzyme-coupling fluorescent method . The glycerol content was determined using a coupled enzyme reaction (Free Glycerol Reagent, Sigma, Cat.# F6428) with the exception that H2O2 production was coupled to peroxidase-dependent formation of the fluorescent dye resorufin. Briefly, rat cerebellar membranes (1 μg/well of 96-well plate), prepared as described in , were pretreated with the serine hydrolase inhibitors MAFP (1 μM) or JZL184 (100 μM) for 30 min and then incubated with or without 18:1 LPA, GP, or (2 S)-OMPT (10 μM final concentration). The fluorescence (λex 530 nm, λem 590 nm) was monitored kinetically for 90 min at 10 min intervals using Tecan Infinite M200 fluorometer.
Extraction of LPA for mass spectrometric determination
Slides with two horizontal brain sections were incubated for 40 min in the presence of AlFx- (NaF 10 mM + AlCl3 50 μM) mimicking the 35 S]GTPγS autoradiography. The control slides were incubated similarly in the assay buffer. After 40 min incubation, slides were rinsed twice (2 min each time) in ice-cold washing buffer, dipped for 30s in ice-cold deionized water and air-dried. The modified extraction method of Bligh and Dyer  was applied for the isolation of LPA from the tissue matrix. One sample consisted of pooled tissue obtained from four slides. The brain tissue was scraped manually from the slides with a spatula using the mixture of 50 mM Tris–HCl, pH 7.40 and methanol with a ratio of 1:4 (v/v); this mixture also included an internal standard (17:0 LPA) used in the quantification. The tissue was transferred to a screw-capped Pyrex® borosilicate glass test tube. The mixture of 50 mM Tris–HCl, pH 7.40 and methanol (1:4, v/v) was added to the test tube to bring the volume up to 200 μl. Chloroform was added to yield a water/methanol/chloroform ratio of 1:4:2 (v/v/v) and the samples were shaken for 1 h with a vertical shaker (Heidolph Multi Reax, Heidolph Instruments GmbH & Co, Schwabach, Germany). To achieve the phase separation, 80 μl of chloroform and 80 μl of water were added. After vortexing for 1 min, the samples were centrifuged at 1800 x g for 15 min at room temperature. The upper aqueous layer was transferred to an HPLC sample vial.
Liquid chromatography/tandem mass spectrometry (LC/MS/MS)
The method for LC/MS/MS determination of LPAs with varying acyl chains has been previously described . The HPLC system comprised of an Agilent 1200 Series Rapid Resolution LC System (Agilent Technologies, Waldbronn, Germany) consisting of a solvent micro vacuum degasser, a binary pump, a thermostatted column compartment SL, and an autosampler SL. Ten microliter of sample solution were injected onto a reversed phase HPLC column (XBridge™ C8 2.1x50 mm, 2.5 μm) (Waters, Ireland) using gradient elution with 50 μM ammonium acetate + 1% triethylamine (TEA) (A) and 1% TEA in 90% methanol (B) as follows: 0–6.0 min 20% B → 90% B, 6.0-10.0 min 90% B, 10.0-10.1 min 90% B → 20% B, 10.0-15.0 min 20% B. The flow rate was 0.3 ml/min, column temperature was maintained at 40 °C and the autosampler tray temperature was set to 10 °C.
The mass spectrometric analysis was carried out with an Agilent 6410 Triple Quadrupole LC/MS equipped with an electrospray ionization source (Agilent Technologies, Palo Alto, CA, USA). The following ionization conditions were used: ESI negative ion mode, drying gas (nitrogen) temperature 300 °C, drying gas flow rate 8 l/min, nebulizer pressure 40 psi and capillary voltage 4000 V. Analyte detection was performed using multiple reaction monitoring (MRM) with the following transitions: m/z 409 → 153 for 16:0 LPA, m/z 437 → 153 for 18:0 LPA, m/z 435 → 153 for 18:1 LPA, m/z 457 → 153 for 20:4 LPA, and m/z 423 → 153 for 17:0 LPA. The fragmentor voltage was 160 V and collision energy 20 V except 23 V for 17:0 LPA. Data were acquired by Agilent MassHunter Workstation Acquisition software (Agilent Technologies, Data Acquisition for Triple Quad., version B.01.03).
Autoradiography films were digitized using a HP scanjet 7400c scanner. For the quantitative data, optical densities on the autoradiograms were measured using ImageJ, a freely available java-based image analysis software system developed in the National Institutes of Health, USA (http://rsb.info.nih.gov/ij/). Optical densities were converted to nCi/g using nonlinear transformation built by the greyscale values of [14 C] standards. In LC/MS/MS experiments, an internal standard (17:0 LPA) was used for quantification, and peak area ratios of the analyte to the IS were calculated as a function of the concentration ratios of the analyte to the internal standard using Agilent MassHunter software (Quantitative Analysis Version B.01.03). The protein content of brain sections was determined by the Pierce BCA Protein Assay Kit with BSA as the standard. The statistical differences were determined either using one-way ANOVA with Tukey’s multiple comparison post hoc test or t test (LC/MS/MS experiments) with ***p < 0.001, **p < 0.01, or *p < 0.05 considered as statistically significant. All statistical data analyses were conducted using GraphPad Prism for Windows.
The authors wish to thank M.Sc. Ville Palomäki for his participation in the early stages of this study and for some autoradiography images used in this manuscript. We wish to acknowledge Ms. Pirjo Hänninen, Ms. Satu Marttila, Ms. Taija Hukkanen, and Ms. Minna Glad for their highly competent technical assistance in the laboratory, and Dr. Ewen MacDonald for revising the language of this manuscript. We also thank Ms. Elena Fonalleras Lozano and Ms. Casandra Riera Ribas for performing additional experiments relevant to this paper as Erasmus exchange students. The MAG lipase part of this work (depicted in Figure 7) was supported by the Academy of Finland (grant 139620 to JTL).
- 10.Nakane S, Oka S, Arai S, Waku K, Ishima Y, Tokumura A, Sugiura T: 2-Arachidonoyl-sn-glycero-3-phosphate, an arachidonic acid-containing lysophosphatidic acid: occurrence and rapid enzymatic conversion to 2-arachidonoyl-sn-glycerol, a cannabinoid receptor ligand, in rat brain. Arch Biochem Biophys. 2002, 402: 51-58.CrossRefPubMedGoogle Scholar
- 11.Sugiura T, Nakane S, Kishimoto S, Waku K, Yoshioka Y, Tokumura A, Hanahan DJ: Occurrence of lysophosphatidic acid and its alkyl ether-linked analog in rat brain and comparison of their biological activities toward cultured neural cells. Biochim Biophys Acta. 1999, 1440: 194-204.CrossRefPubMedGoogle Scholar
- 16.Brindley DN, Pilquil C: Lipid phosphate phosphatases and signaling. J Lipid Res. 2009, 50 (Suppl): 225-230.Google Scholar
- 35.Simon MF, Rey A, Castan-Laurel I, Gres S, Sibrac D, Valet P, Saulnier-Blache JS: Expression of ectolipid phosphate phosphohydrolases in 3T3F442A preadipocytes and adipocytes. Involvement in the control of lysophosphatidic acid production. J Biol Chem. 2002, 277: 23131-23136.PubMedCentralCrossRefPubMedGoogle Scholar
- 43.Long JZ, Li W, Booker L, Burston JJ, Kinsey SG, Schlosburg JE, FJ Pavo´ n, Serrano AM, Selley DE, Parsons LH, Lichtman AH, Cravatt BF: Selective blockade of 2-arachidonoylglycerol hydrolysis produces cannabinoid behavioral effects. Nat Chem Biol. 2009, 5: 37-44.PubMedCentralCrossRefPubMedGoogle Scholar
- 52.Trimbuch T, Beed P, Vogt J, Schuchmann S, Maier N, Kintscher M, Breustedt J, Schuelke M, Streu N, Kieselmann O, Brunk I, Laube G, Strauss U, Battefeld A, Wende H, Birchmeier C, Wiese S, Sendtner M, Kawabe H, Kishimoto-Suga M, Brose N, Baumgart J, Geist B, Aoki J, Savaskan NE, Bräuer AU, Chun J, Ninnemann O, Schmitz D, Nitsch R: Synaptic PRG-1 modulates excitatory transmission via lipid phosphate-mediated signaling. Cell. 2009, 138: 1222-1235.PubMedCentralCrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.